Microbiology

Published by Society for General Microbiology
Online ISSN: 1465-2080
Print ISSN: 1350-0872
Publications
(a) Detection of the potential pGTF gene by PCR using hybrid primers: lane 1, B. longum ATCC 15707 T ( B. longum subsp. longum type strain producing EPS); lane 2, B. longum subsp. longum CRC 002; M, 100 bp DNA ladder. (b) Hydrophobicity plot of the WblA putative protein product using the TMpred program. The preferred model predicts that the N- terminus is inside, followed by six strong transmembrane helices. (c) CRC 002 chromosomal DNA restriction map of the putative EPS biosynthesis region. Restriction fragments are designated by numbers indicating the sizes in kb and letters indicating the restriction sites (E, Eco RI; H, Hin dIII; K, Kpn I; X, Xho I). (d) Predicted ORFs (arrows) from the sequenced region; gene designations are indicated below the arrows. Potential functions are based on homology searches as described in Methods. 
Comparison of the sequenced region from B. longum subsp. longum CRC 002 with B. longum subsp. longum NCC 2705 (GenBank accession no. AE014295), B. longum subsp. longum DJO10A (CP000605) and B. longum subsp. infantis CCUG 52486 (DS990238). Arrows with the same colour share the same putative function. 
Growth profiles of B. longum subsp. longum CRC 002 in MRSCG2 (a), MRSCGA2 (b), MRSCF2 (c) and MRSCL2 (d) at 37 6C. Viable cell count ($, solid lines), pH (m; dotted lines) and titratable acidity in % of equivalent lactic acid (X; dashed lines). Error bars: standard error of four independent experiments.
The effect of four sugars (glucose, galactose, lactose and fructose) on exopolysaccharide (EPS) production by Bifidobacterium longum subsp. longum CRC 002 was evaluated. More EPS was produced when CRC 002 was grown on lactose in the absence of pH control, with a production of 1080+/-120 mg EPS l(-1) (P<0.01) after 24 h of incubation. For fructose, galactose and glucose, EPS production was similar, at 512+/-63, 564+/-165 and 616+/-93 mg EPS l(-1), respectively. The proposed repeating unit composition of the EPS is 2 galactose to 3 glucose. The effect of sugar and fermentation time on expression of genes involved in sugar nucleotide production ( galK, galE1, galE2, galT1, galT2, galU, rmlA, rmlB1 and rmlCD) and the priming glycosyltransferase ( wblE) was quantified using real-time reverse transcription PCR. A significantly higher transcription level of wblE (9.29-fold) and the genes involved in the Leloir pathway (galK, 4.10-fold; galT1, 2.78-fold; and galE2, 4.95-fold) during exponential growth was associated with enhanced EPS production on lactose compared to glucose. However, galU expression, linking glucose metabolism with the Leloir pathway, was not correlated with EPS production on different sugars. Genes coding for dTDP-rhamnose biosynthesis were also differentially expressed depending on sugar source and growth phase, although rhamnose was not present in the composition of the EPS. This precursor may be used in cell wall polysaccharide biosynthesis. These results contribute to understanding the changes in gene expression when different sugar substrates are catabolized by B. longum subsp. longum CRC 002.
 
In Streptomyces sp. FR-008, the biosynthetic gene cluster of the polyene antibiotic FR-008, also known as candicidin, consists of twenty-one genes, including four regulatory genes, fscRI to fscRIV. Our bioinformatics analyses indicate that FscRI has an N-terminal PAS domain, whereas the other three regulators have N-terminal AAA domains and are members of the LAL (large ATP-binding regulators of the LuxR type) family. Deletion of fscRI abolished the production of FR-008, with production restored in the complemented strain, supporting a critical role for FscRI in FR-008 biosynthesis. Consistent with these findings, transcription of genes involved in the biosynthesis and efflux of FR-008 was greatly downregulated in ΔfscRI. Interestingly, the regulatory gene fscRIV was also downregulated in ΔfscRI. Production of FR-008 was reduced, but not abrogated, in an fscRIV deletion mutant, and although structural genes were downregulated in ΔfscRIV, the changes were much less dramatic than in ΔfscRI, suggesting a stronger regulatory role for FscRI. Remarkably, transcription of fscRI was also decreased in ΔfscRIV. Expression of fscRI restored antibiotic production in an ΔfscRIV mutant, but not vice versa. Putative binding sequences for FscRI were identified upstream of fscRIV and the three structural genes fscA, fscB and fscD, which encode large modular PKSs. Our findings suggest that fscRI and fscRIV are interregulatory, whereas expression of fscRII and fscRIII appeared to be independent of fscRI and fscRIV. This study demonstrates that the regulation of polyene antibiotic synthesis can involve mutually regulated transcriptional activators that belong to different families.
 
Representative consensus neighbourjoining tree of the catalytic domain (left of figure), and schematic diagrams of the domain structure (right of figure) of MCP-01 and related sequences. The consensus tree of the catalytic domain (peptidase_S8) was constructed using MEGA 3.1 with PAM model. One thousand bootstrap replicates were used. Only bootstrap percentages .40 % are shown on the branches. The conserved domain architectures of the sequences were identified using the CD-search service available at NCBI (http://www.ncbi.nlm.nih.gov/Structure/cdd/ cdd.shtml). The conserved domains are: peptidase_S8, subtilase family, pfam00082; subtilisin_N, subtilisin N-terminal region, pfam05922; PPC, bacterial pre-peptidase Cterminal domain, pfam04151; P_proprotein, proprotein convertase P-domain, pfam01483; PKD, PKD I domain, cd00146 (or smart00089). The deseasin MCP-01 sequence is indicated by an asterisk. Detailed information about the sequences used for tree construction is shown in Supplementary Table S1. 
Ribbon representation of the catalytic domain (a) and the PKD domain (b) of MCP01. Homology modelling was conducted at CPHmodels 2.0 Server (http://www.cbs.dtu. dk/services/CPHmodels/) (Lund et al., 2002) with templates 1DBI.pdb for the catalytic domain and 1R64.pdb for the PKD domain. The secondary structural elements are represented in yellow (helices), red (sheets) and cyan (random coils), and the catalytic triad in green. Two segments of the loops shown in blue are the inserts in MCP-01 compared to other subfamily S8A subtilases. The figure was prepared with Swiss-PdbViewer version 3.7 (Guex & Peitsch, 1997). 
Sequence alignment of MCP-01 and subtilases in family S8. Asp, His, and Ser of the catalytic triad are shown in bold type. The positions conserved in all six sequences are boxed. For each sequence, the beginning and end positions in the original sequence (the corresponding gi number is shown below) are indicated in brackets. Sequence positions included in the phylogenetic analysis are indicated by an asterisk in the bottom row. Abbreviations: Sub.Carls., subtilisin Carlsberg, Bacillus licheniformis, gi135016; Thermitase, thermitase, Bacillus cereus E33L, gi52143098; Sub.Ak.1, subtilisin Ak.1, Bacillus sp. PD498, gi1890101; Peptidase K, peptidase K, Tritirachium album, gi131077; Deseasin, deseasin MCP-01, Pseudoalteromonas sp. SM9913, gi110083909; Kexin, kexin, Saccharomyces cerevisiae, gi125350. 
MCP-01, the main protease secreted by the deep-sea cold-adapted bacterium Pseudoalteromonas sp. SM9913, is a cold-adapted serine protease. Gene mcp01 encoding MCP-01 contains an ORF of 2508 bp encoding a protein of 835 amino acid residues with an M(r) of 87 773 Da, which is a multidomain subtilase precursor. Mature MCP-01 purified from the culture of strain SM9913 with an M(r) of 65.84 kDa is a multidomain protein composed of a catalytic domain, a linker, a P_proprotein domain and a polycystic kidney disease (PKD) domain. To the best of the authors' knowledge, no mature subtilase has been reported to date with this domain architecture. Phylogenetic analyses of subtilases showed that MCP-01 and 12 hypothetical proteins retrieved from public databases form a strongly supported group within the subtilase subfamily. These 13 proteins are predicted to share a similar domain architecture and represent a structurally novel group within the S8A subfamily. The substrate specificities of MCP-01 towards synthetic peptides differed from that of a typical S8A protease, subtilisin Carlsberg. Since most of this new subgroup of subtilases, including MCP-01 and the 12 MCP-01-like subtilases, are from deep-sea bacteria, they are termed deseasins. MCP-01 is the type example of a deseasin, since it is the only one that has been purified and characterized. In addition, the structural characteristics and catalytic properties of deseasin MCP-01 show that structurally and kinetically it is adapted to low temperatures.
 
We have identified a gene, vlpA, which is closely linked to the mfrA,B locus associated with mannose-fucose-resistant haemagglutination. VlpA is an outer-membrane protein which can be labelled with [3H]palmitate and whose processing is globomycin-sensitive, suggesting that it is a lipoprotein. Homology searches revealed that VlpA belongs to the group of lipocalins of the alpha 2-microglobulin superfamily which function as small hydrophobic molecule transporters, and is the first identified bacterial member of this group. Multiple copies of this gene are present in Vibrio cholerae O1 and O139 and Southern hybridization reveals a biotype-specific pattern of fragment sizes. Construction of strains capable of hyperproducing VlpA suggested that it is able to bind haemin with low affinity but this may be due to a simple hydrophobic interaction. Attempts to construct specific mutants in vlpA have been unsuccessful, presumably because of the multiple copies of vlpA genes and their linkage to the VCR element.
 
Copper homeostasis in V. cholerae involves the coordinated action of CopA, CueR and the periplasmic proteins Cot and CopG. Question marks indicate that the mechanism used to transport copper inside the cell is unknown, as is that used by Cot and CopG to detoxify copper.
The bacterial pathogen Vibrio cholerae requires colonizination of the human small intestine to cause cholera. The anaerobic and slightly acidic conditions predominating there enhance toxicity of low copper concentrations and create a selective environment for bacteria with evolved detoxifying mechanisms. We reported previously that the VCA0260, VCA0261 and VC2216 gene products were synthesized only in V. cholerae grown in microaerobiosis or anaerobiosis. Here we show that ORFs VCA0261 and VCA0260 are actually combined into a single gene encoding a 18.7 kDa protein. Bioinformatic analyses linked this protein and the VC2216 gene product to copper tolerance. Following the approach of predict-mutate and test, we describe for the first time, to our knowledge, the copper tolerance systems operating in V. cholerae. Copper susceptibility analyses of mutants in VCA0261-0260, VC2216 or in the putative copper-tolerance-related VC2215 (copA ATPase) and VC0974 (cueR), under aerobic and anaerobic growth, revealed that CopA represents the main tolerance system under both conditions. The VC2216-encoded periplasmic protein contributes to resistance only under anaerobiosis in a CopA-functional background. The locus tag VCA0261-0260 encodes a copper-inducible, CueR-dependent, periplasmic protein, which mediates tolerance in aerobiosis, but under anaerobiosis its role is only evident in CopA knock-out mutants. None of the genes involved in copper homeostasis were required for V. cholerae virulence or colonization in the mouse model. We conclude that copper tolerance in V. cholerae, which lacks orthologues of the periplasmic copper tolerance proteins CueO, CusCFBA and CueP, involves CopA and CueR proteins along with the periplasmic Cot (VCA0261-0260) and CopG (VC2216) V. cholerae homologues.
 
Streptococcus suis serotype 2 (S. suis 2) is an important zoonotic pathogen. It causes heavy economic losses in the pig-farming industry and can be associated with severe infections in humans, e.g. streptococcal toxic shock syndrome. Understanding its pathogenesis is critical for prevention and control of diseases caused by S. suis 2. In this study, we show that deletion of a two-component system (TCS), 05SSU1660/1661 (orthologues of the Ihk/Irr TCS of Streptococcus pyogenes), in S. suis 2 strain 05ZYH33 results in notable attenuation of virulence, as exemplified by reduced adherence to mucosal epithelium cells, increased elimination by macrophages, reduced ability to survive in an acidic or oxidative-stressed environment, and lowered pathogenicity in mice. Further analysis of differential proteomics profiles by two-dimensional electrophoresis revealed that while many previously well-known virulence factors, such as suilysin, autolysin and muraminidase-released protein, were not expressed differentially, cell metabolism was downregulated in the Ihk/Irr deletion mutant. In addition, the oxidative-stress response gene for manganese-dependent superoxide dismutase (MnSOD) was also repressed significantly in the mutant. Collectively, our data suggest that the Ihk/Irr TCS contributes to the virulence of S. suis 2 strain 05ZYH33, mainly through alteration of the bacterial cell metabolism.
 
SDS-PAGE of water-soluble proteins extracted from Synechococcus sp. PCC 7942 (lanes 1 and 2) and Oscillatoria sp. NKBG 091600 (lanes 3 and 4) irradiated with UV-A. Lanes 1 and 3, cells were irradiated with a cool-white fluorescent light; lanes 2 and 4, cells were irradiated with a combination of coolwhite fluorescent light and UV-A light for 15 h. M, size markers. The induced 60 kDa protein in lane 4 is indicated by an arrowhead. 
Protein profile of water-soluble fraction (a) and Northern blot of groEL transcripts (b) in Oscillatoria sp. NKBG 091600 during 900 min UV-A irradiation. M, molecular mass markers (kDa). 
The putative groESL operon transcriptional start site determined by primer extension of RNA from cells irradiated with UV-A. An oligonucleotide complementary to groESL mRNA was used for both reverse transcriptase primer extension (farright lane) and nucleotide sequencing. The DNA sequence of the putative SO5 box is side-lined. + 1, putative transcriptional start site. 
The authors have examined the response to UV-A irradiation of the UV-A-resistant marine cyanobacterium Oscillatoria sp. NKBG 091600, which produces the UV-A-absorbing compound biopterin glucoside. The expression of a 60 kDa protein was markedly induced at 500 min after UV-A irradiation. This protein was identified by N-terminal amino acid sequence analysis as GroEL. Northern blot analysis demonstrated that GroEL synthesis was controlled by UV-A at the transcriptional level. A CIRCE element and a putative SOS consensus sequence were found upstream of the groESL operon, overlapping two putative promoter sequences. Primer extension analysis revealed that groESL transcription in UV-A-induced cells starts from the proximal promoter overlapped by the SOS consensus sequence. This indicates that an SOS response regulation is instrumental in UV-A-induced GroEL expression of Oscillatoria sp. NKBG 091600. Furthermore, this UV-A-inducible GroEL may function to upregulate biopterin glucoside biosynthesis, thereby allowing growth under UV-A irradiation.
 
A novel carotenoid 1,2-hydratase (CruF) responsible for the C-1',2' hydration of gamma-carotene was identified in the non-photosynthetic bacteria Deinococcus radiodurans R1 and Deinococcus geothermalis DSM 11300. Gene expression and disruption experiments demonstrated that dr0091 and dgeo2309 encode CruF in D. radiodurans and D. geothermalis, respectively. Their homologues were also found in the genomes of cyanobacteria, and exhibited little homology to the hydroxyneurosporene synthase (CrtC) proteins found mainly in photosynthetic bacteria. Phylogenetic analysis showed that CruF homologues form a separate family, which is evolutionarily distant from the known CrtC family.
 
Effect of potassium glutamate on the incorporation of [ 14 C]glucose into glucan-protein intermediates of different strains. Total membranes (0n2 mg protein) of different strains were incubated with UDP-[ 14 C]glucose as described in Methods, with or without the addition of 0n5 M potassium glutamate. Reactions were stopped by the addition of 10 % TCA and the precipitates subjected to SDS-PAGE. Radioactivity was detected by autoradiography. (a), A. tumefaciens A348 ; (b) R. meliloti 102F34 ; (c), B. abortus S19 ; (d), R. meliloti GR4 ; (e), A. tumefaciens A1011(pBA19) ; (f), R. meliloti GRT21s(pBA19) ; (g), R. meliloti GRT21s(pCD523) ; (h), B. abortus BAI129(pCD523). 
Bio-Gel P4 chromatography of cyclic 1,2-β-glucans accumulated in vivo by different strains in cells grown in medium of low and high osmolarity. Cells from 100 ml cultures were harvested and extracted with 10 % TCA. Extracts were subjected to Bio-Gel P4 chromatography. Fractions were collected and carbohydrates measured as indicated in Methods. Carbohydrates were expressed as mg glucose equivalents (g wet weight) − 1. (a), A. tumefaciens A348 ; (b), A. tumefaciens A1011(pBA19) ; (c), B. abortus S19 ; (d), B. abortus BAI129(pCD523) ; (e), R. meliloti GR4 ; (f), R. meliloti GRT21s(pCD523) ; (g), R. meliloti GRT21s(pBA19). (a), (b), (e), (f) and (g) contained 0n25 M NaCl ; (c) and (d) contained 0n5 M mannitol. $, control ; #, with 0n25 M NaCl or 0n5 M mannitol. 
DEAE-Sephadex chromatography of cellular glucans accumulated by different strains. Glucans were recovered from Bio-Gel P4 columns (fractions 5 to 18 of Fig. 2) and subjected to DEAE-Sephadex chromatography. Columns (0n25i3 cm) were eluted with 1n5 ml water (fractions 1-3), 1n5 ml 0n25 M NaCl (fractions 4-6) and 0n5 M NaCl (fractions 7-9). Fractions of 0n5 ml were collected and carbohydrates measured by the anthrone-sulfuric acid method (Dische et al., 1962). (a), R. meliloti GR4 grown in AMA medium ; (b), R. meliloti GRT21s(pBA19) grown in AMA medium ; (c), R. meliloti GRT21s(pBA19) grown in AMA medium with 0n25 M NaCl ; (d), A. tumefaciens A348 grown in TY medium ; (e), A. tumefaciens A1011(pBA19) grown in TY medium ; (f), A. tumefaciens A1011(pBA19) grown in TY medium with 0n25 M NaCl. 
In contrast to what happens in Agrobacterium tumefaciens and Rhizobium meliloti, synthesis of periplasmic cyclic 1,2-β-glucan in Brucella spp. was not inhibited when bacteria were grown in media of high osmolarity. Studies performed with crude membrane preparations showed that cyclic 1,2-β-glucan synthetase of Brucella spp. was not inhibited by 0·5 M KCI or potassium glutamate; concentrations that completely inhibit the osmosensitive enzymes of A. tumefaciens A348 or R. meliloti 102F34, respectively encoded by the chvB or ndvB genes. The Brucella abortus cyclic 1,2-β-glucan synthetase gene (cgs) was introduced into A. tumefaciens A1011 chvB and R. meliloti GRT21s ndvB mutants. Synthesis of cyclic 1,2-β-glucan by the recombinant strains was not inhibited when grown in media of high osmolarity (0·25 M NaCl or 0·5 M mannitol). On the other hand, when the A. tumefaciens cyclic 1,2-β-glucan synthetase gene was introduced into the R. meliloti GRT21s ndvB mutant, the recombinant strain displayed marked inhibition of cyclic 1,2-β-glucan synthesis when grown in high-osmolarity media. However, the same gene introduced into a B. abortus cgs mutant background resulted in no inhibition of glucan synthesis at high osmolarity. In vitro studies with crude membranes isolated from recombinant strains revealed that Brucella cyclic 1,2-β-glucan synthetase was not inhibited by high concentrations of KCI or potassium glutamate even when expressed in Agrobacterium or Rhizobium backgrounds. It was concluded that the lack of effect of high osmolarity on 1,2-β-glucan synthesis in Brucella is due to two convergent mechanisms: a) the presence of a cyclic 1,2-β-glucan synthetase that is not affected by concentrations of solutes such as KCI or potassium glutamate and b) either the possible accumulation of compatible solutes that might protect the enzyme from the inhibition by potassium glutamate or the accumulation of other osmolytes that do not affect the 1,2-β-glucan synthetase.
 
The transfer of phosphoglycerol moieties from phosphatidylglycerol to the cyclic (1,2)-beta-glucans in growing cultures of Rhizobium meliloti strain 1021 was investigated using pulse-chase experiments with [3H]glycerol and/or [14C]glucose. No transfer occurred when cells were grown and pulse-chased in a medium containing 0.4 M NaCl. However, radiolabelled glycerophosphorylated cyclic (1,2)-beta-glucans could be detected within 30 min after transfer of these cultures to a low-osmolarity medium. Conversely, when low-osmolarity cultures were shifted to a high-osmolarity medium containing 0.4 M NaCl or 0.8 M sucrose, the transfer of phosphoglycerol substituents to the cyclic (1,2)-beta-glucans was inhibited. Further experiments revealed that the transfer of phosphoglycerol substituents to the cyclic (1,2)-beta-glucans occurs within the periplasmic compartment.
 
P. naphthalenivorans strain CJ2 genes and gene products
Polaromonas naphthalenivorans strain CJ2 metabolizes naphthalene via the gentisate pathway and has recently been shown to carry a third copy of gentisate 1,2-dioxygenase (GDO), encoded by nagI3, within a previously uncharacterized naphthalene catabolic gene cluster. The role of this cluster (especially nagI3) in naphthalene metabolism of strain CJ2 was investigated by documenting patterns in regulation, transcription and enzyme activity. Transcriptional analysis of wild-type cells showed the third cluster to be polycistronic and that nagI3 was expressed at a relatively high level. Individual knockout mutants of all three nagI genes were constructed and their influence on both GDO activity and cell growth was evaluated. Of the three knockout strains, CJ2ΔnagI3 showed severely diminished GDO activity and grew slowest on aromatic substrates. These observations are consistent with the hypothesis that nagI3 may prevent toxic intracellular levels of gentisate from accumulating in CJ2 cells. All three nagI genes from strain CJ2 were cloned into Escherichia coli: the nagI2 and nagI3 genes were successfully overexpressed. The subunit mass of the GDOs were ~36-39 kDa, and their structures were deduced to be dimeric. The K(m) values of NagI2 and NagI3 were 31 and 10 µM, respectively, indicating that the higher affinity of NagI3 for gentisate may protect the wild-type cells from gentisate toxicity. These results provide clues for explaining why the third gene cluster, particularly the nagI3 gene, is important in strain CJ2. The organization of genes in the third gene cluster matched that of clusters in Polaromonas sp. JS666 and Leptothrix cholodnii SP-6. While horizontal gene transfer (HGT) is one hypothesis for explaining this genetic motif, gene duplication within the ancestral lineage is equally valid. The HGT hypothesis was discounted by noting that the nagI3 allele of strain CJ2 did not share high sequence identity with its homologues in Polaromonas sp. JS666 and L. cholodnii SP-6.
 
Cytosolic alpha-mannosidases are glycosyl hydrolases that participate in the catabolism of cytosolic free N-oligosaccharides. Two soluble alpha-mannosidases (E-I and E-II) belonging to glycosyl hydrolases family 47 have been described in Candida albicans. We demonstrate that addition of pepstatin A during the preparation of cell homogenates enriched alpha-mannosidase E-I at the expense of E-II, indicating that the latter is generated by proteolysis during cell disruption. E-I corresponded to a polypeptide of 52 kDa that was associated with mannosidase activity and was recognized by an anti-alpha1,2-mannosidase antibody. The N-mannan core trimming properties of the purified enzyme E-I were consistent with its classification as a family 47 alpha1,2-mannosidase. Differential density-gradient centrifugation of homogenates revealed that alpha1,2-mannosidase E-I was localized to the cytosolic fraction and Golgi-derived vesicles, and that a 65 kDa membrane-bound alpha1,2-mannosidase was present in endoplasmic reticulum and Golgi-derived vesicles. Distribution of alpha-mannosidase activity in a kex2Delta null mutant or in wild-type protoplasts treated with monensin demonstrated that the membrane-bound alpha1,2-mannosidase is processed by Kex2 protease into E-I, recognizing an atypical cleavage site of the precursor. Analysis of cytosolic free N-oligosaccharides revealed that cytosolic alpha1,2-mannosidase E-I trims free Man8GlcNAc2 isomer B into Man7GlcNAc2 isomer B. This is believed to be the first report demonstrating the presence of soluble alpha1,2-mannosidase from the glycosyl hydrolases family 47 in a cytosolic compartment of the cell.
 
IFA of C. albicans W 3 2 grown in RPMl medium containing 10 pg tunicamycin ml-'. (a) Labelling with mAb AF1 specific for P-1,2-oligomannosidic epitopes was restricted to mother cells from the inoculum whilst young buds (arrows) were unlabelled. (b) Autoradiography of French press extracts metabolically labelled with [2-3H]mannose in the absence (lane 1) or presence (lane 2) of 10 yg tunicamycin ml-'. The PLM nature of the material labelled in the 14-18 kDa range was confirmed in a chloroform/methanol/water extract (lane 3).  
IFA of C. albicans W 3 2 grown in yeast extract-Sabouraud broth a t pH 5.6 (a) or pH 2 (b). Labelling with mAb AF1 in (a) revealed expression of ~-1,2-oligomannosidic epitopes whilst cells in (b) were unlabelled. Western blots of AERC extracts performed on cells grown either a t pH 5.6 (c, lanes 1 and 2) or a t pH 2 (c, lanes 3 and 4) revealed the presence of PLM in both cases. Strips were probed either with ConA-peroxidase (lanes 1 and 3) or mAb AF1 (lanes 2 and 4).  
Fig. l a Differential distribution of a-linked (a) and /3-linked (b) mannose epitopes a t the cell surface of C. d/bicdns VW32 grown in RPMI medium a t 37°C as determined by a double immunofluorescence assay. (a) mAb CA1 binding revealed by a fluorescein-goat anti-rat IgM conjugate; (b) mAb AF1 binding revealed with a Texas red-goat anti-mouse IgM conjugate.  
Western blots of AERC extracts from mannan mutant strains KD 101 (lane l), A4 (lane 2) and I F 0 1397 (lane 3) grown on SDA. Strips were probed with mAb AF1.  
A monoclonal antibody specific for beta-1,2-linked oligomannosides was used to study the association of these residues with Candida albicans mannan and phospholipomannan (PLM) in relation to growth conditions and in mannan mutant strains. Double immunofluorescence assays performed on cells grown under standard conditions indicated a highly heterogeneous cell surface expression of these epitopes in comparison with the homogeneous expression of alpha-linked oligomannosidic epitopes. Growth in the presence of tunicamycin, which inhibits mannan N-glycosylation, resulted in an absence of beta-1,2-oligomannosidic epitopes on the cell surface, although PLM synthesis still occurred as shown by autoradiography. Similarly, growth in acidic conditions, which inhibits the incorporation of beta-1,2-oligomannosides in mannan, resulted in an absence of beta-1,2-oligomannosidic epitopes at the cell surface, although they still associated with PLM as shown by Western blotting. Western blots of C. albicans mutant strains with reduced amounts or an absence of phosphorus and acid-labile beta-1,2-oligomannosides in their mannan confirmed that the association of beta-1,2-linked oligomannosides with mannan and with PLM involves different mannosylation processes.
 
Western blots of AERC extracts of yeast cells of C. albicans W 32 stained with ConA (lane l), serum factor 5 (lane 2), mAb DF9-3 (lane 3) or mAb AF1 (lanes 4 and 5). Yeast cells were grown either on SDA (lanes 1-4) or in yeast extract broth in a bioreactor (lane 5). 
The distribution of beta-1,2-linked oligomannosides among glycoconjugates of various Candida species was investigated by Western blotting, using monoclonal and polyclonal antibodies which react with these epitopes. Expression of beta-1,2-linked oligomannosidic epitopes on a 14-18 kDa polydisperse antigen nonreactive with concanavalin A (ConA), previously identified as a C. albicans serotype A phospholipomannan (PLM), appeared to be restricted to C. albicans serotypes A and B (including var. C. stellatoidea types I and II) and C. tropicalis. In C. albicans, beta-1,2-linked oligomannosidic epitopes also appeared to be slightly associated with high molecular mass (> 100 kDa) polydisperse ConA-reactive mannoproteins. For all the other Candida strains investigated, belonging to the species C. parapsilosis, C. krusei, C. glabrata and C. robusta (S. cerevisiae), beta-1,2-linked oligomannosidic epitopes were found to be present in association with medium molecular mass (18-100 kDa) and high molecular mass ConA-reactive mannoproteins, giving reproducible labelling profiles that varied between species.
 
Proteome analysis of bacteria that can detoxify harmful organic compounds enables the discovery of enzymes involved in the biodegradation of these substances and proteins that protect the cell against poisoning. Exposure of Delftia acidovorans MC1 to 2,4-dichlorophenoxypropionic acid and its metabolites 2,4-dichlorophenol and 3,5-dichlorocatechol during growth on pyruvate as a source of carbon and energy induced several proteins. Contrary to the general hypothesis that lipophilic or reactive compounds induce heat shock or oxidative stress proteins, no induction of the GroEL, DnaK and AhpC proteins that were used as markers for the induction of heat shock and oxidative stress responses was observed. However, two chlorocatechol1,2-dioxygenases, identified by amino terminal sequence analysis, were induced. Both enzymes catalyse the conversion of 3,5-dichlorocatechol to 2,4-dichloro-cis,cis-muconate indicating that biodegradation is a major mechanism of resistance in the detoxifying bacterium D. acidovorans MC1.
 
Sporothrix (Sp.) schenckii is a pathogenic fungus that infects humans and animals, and is responsible for the disease named sporotrichosis. The cell wall of this fungus has glycoproteins with a high content of mannose and rhamnose units, which are synthesized by endoplasmic reticulum- and Golgi-localized glycosyltransferases. Little is known about the enzymic machinery involved in the synthesis of these oligosaccharides in Sp. schenckii, or the genes encoding these activities. This is in part because of the lack of an available genome sequence for this organism. Using a partial genomic DNA library we identified SsMNT1, whose predicted product has significant similarity to proteins encoded by members of the Saccharomyces (Sa.) cerevisiae KRE2/MNT1 gene family. In order to biochemically characterize the putative enzyme, SsMNT1 was heterologously expressed in the methylotrophic yeast Pichia pastoris. Recombinant SsMnt1 showed Mn(2+)-dependent mannosyltransferase activity and the ability to recognize as acceptors α-methyl mannoside, mannose, Man(5)GlcNAc(2) oligosaccharide and a variety of mannobiosides. The characterization of the enzymic products generated by SsMnt1 revealed that the enzyme is an α1,2-mannosyltransferase that adds up to two mannose residues to the acceptor molecule. Functional complementation studies were performed in Sa. cerevisiae and Candida albicans mutants lacking members of the KRE2/MNT1 gene family, demonstrating that SsMnt1 is involved in both the N- and O-linked glycosylation pathways, but not in phosphomannan elaboration.
 
PAGE of total membranes and permeabilized cells incubated with UDP-['4C]Glc. Lanes: 1 and 2, inner membranes (0.2 mg protein) of A. tumefaciens A348; 3, 4 and 5, permeabilized cells (0-2 mg protein) of B. abortus 519; 6, 7 and 8 permeabilized cells (0.2 mg protein) of B. ovis RE0198; 1, 3 and 6, 10 min incubation with UDP-[14C]Glc; 2, 4 and 7, chase experiment in which, after 10 min incubation with UDP-[14C]Glc, 2 mM non-radioactive UDP-Glc was added and the incubations were continued for 30 min; 5 and 8, chase experiment with 20mM UDP-Glc. Proteins were stained with Coomassie blc;e (a) and radioactivity, detected by fluorography (b). 
PAGE of membranes incubated with UDP-['4C]Glc. Total membranes (0.2 mg protein) of B. ovis strain RE0198 (lanes 3 and 4) or inner membranes (0.2 mg protein) of A. tumefaciens A348 (lanes 1 and 2 ) or R. fredii USDA191 (lanes 5 and 6) were incubated with UDP-[14C]Glc, the reactions were stopped by the addition of 10% TCA and the precipitates were subjected to gel electrophoresis. Proteins were stained with Coomassie blue (a) and radioactivity, detected by fluorography (b). For the chase experiment (even-numbered lanes), 2 mM non-radioactive UDP-Glc was added after a 10min incubation and the reactions were stopped after 30 min. Numbers on the right indicate the molecular masses of standards. 
Biosynthesis of periplasmic cyclic 1,2-beta-glucans in Brucella ovis strain REO198 and B. abortus strain 519 was found to be carried out by membrane-bound enzymes that use UDP-glucose (UDP-Glc) as donor substrate. Contrary to what happens in species of the genera Agrobacterium and Rhizobium, the accumulation of the reaction products in Brucella appeared not to be osmotically regulated. Incubation of permeabilized cells with UDP-[14C]Glc led to the formation of soluble neutral cyclic 1,2-beta-glucans and [14C]glucose-labelled glucoproteins. PAGE of pulse-chase experiments carried out with permeabilized cells showed that the molecular mass of the labelled protein was indistinguishable from Agrobacterium tumefaciens A348 and Rhizobium fredii USDA191 glucoproteins known to be intermediates in the synthesis of cyclic glucans. Brucella total membrane preparations were less efficient than permeabilized cells in the formation of cyclic glucan; this was attributed to defective cyclization. Accumulation of protein intermediates having oligosaccharides of high molecular mass that were not released from the protein was observed after chase with 2 mM UDP-Glc. This defect was not observed when permeabilized cells were used as enzyme preparation, thus suggesting that in Brucella a factor(s) that was lost or inactivated upon the preparation of membranes was required for the effective regulation between elongation and cyclization reactions.
 
The pathogenic fungus Candida albicans is able to cover its most potent proinflammatory cell wall molecules, the β-glucans, underneath a dense mannan layer, so that the pathogen becomes partly invisible for immune cells such as phagocytes. As the C. albicans histidine kinases Chk1p, Cos1p and CaSln1p had been reported to be involved in virulence and cell wall biosynthesis, we investigated whether deletion of the respective genes influences the activity of phagocytes against C. albicans. We found that among all histidine kinase genes, CHK1 plays a prominent role in phagocyte activation. Uptake of the deletion mutant Δchk1 as well as the acidification of Δchk1-carrying phagosomes was significantly increased compared with the parental strain. These improved activities could be correlated with an enhanced accessibility of the mutant β-1,3-glucans for immunolabelling. In addition, any inhibition of β-1,3-glucan-mediated phagocytosis resulted in a reduced uptake of Δchk1, while ingestion of the parental strain was hardly affected. Moreover, deletion of CHK1 caused an enhanced release of interleukins 6 and 10, indicating a stronger activation of the β-1,3-glucan receptor dectin-1. In conclusion, the Chk1p protein is likely to be involved in masking β-1,3-glucans from immune recognition. As there are no homologues of fungal histidine kinases in mammals, Chk1p has to be considered as a promising target for new antifungal agents.
 
Genes encoding Bacillus amyloliquefaciens (1,3-1,4)-beta-glucanase (AMY), B. macerans (1,3-1,4)-beta-glucanase (MAC), and a series of hybrid enzymes containing N-terminal sequence segments of different length derived from AMY with the remaining C-terminal segment derived from MAC, were expressed in Saccharomyces cerevisiae. The cells secreted active enzyme into the medium. While the quantity of N-glycan linked to the different enzymes was similar, pronounced differences in thermotolerance were observed when the glycosylated enzymes were compared with the unglycosylated counterparts produced in Escherichia coli. Glycosylated AMY and hybrid enzyme H(A16-M), consisting of 16 N-terminal amino acids derived from AMY with the remaining C-terminal segment from MAC, exhibited a 7.5- and 1.6-fold increase in half-life at 70 degrees C, pH 6.0. N-terminal sequencing established that only two out of three sites for potential N-glycosylation of H(A16-M) secreted from yeast were actually glycosylated. Removal of N-glycans by endoglycosidase H and peptide:N-glycosidase F from H(A16-M) resulted in a 16- and 133-fold decrease of thermostability, demonstrating that N-glycans are a major determinant for the resistance of this enzyme to thermal inactivation. Glycosylated MAC and hybrid enzymes H(A36-M), H(A107-M) and H(A152-M) had increased thermostability but hybrid enzyme H(A78-M) was less thermostable. N-Glycosylation thus changes thermostability of (1,3-1,4)-beta-glucanases with similar primary structure in a variable, so far unpredictable way.
 
Clostridium thermocellum produces one major beta-1,3-glucanase. Genomic DNA fragments containing the gene were cloned from two strains, DSM1237(T) (6848 bp) and F7 (9766 bp). Overlapping sequences were 99.9 % identical. The nucleotide sequences contained reading frames for a putative transposase, endo-beta-1,3-1,4-glucanase CelC, a putative transcription regulator of the LacI type, beta-1,3-glucanase Lic16A and a putative membrane protein. The licA genes of both strains encoded an identical protein of 1324 aa with a calculated molecular mass of 148 kDa. Lic16A is an unusually complex protein consisting of a leader peptide, a threefold repeat of an S-layer homologous module (SLH), an unknown module, a catalytic module of glycosyl hydrolase family 16 and a fourfold repeat of a carbohydrate-binding module of family CBM4a. The recombinant Lic16A protein was characterized as an endo-1,3(4)-beta-glucanase with a specific activity of 2680 and 340 U mg(-1) and a K(m) of 0.94 and 2.1 mg ml(-1) towards barley beta-glucan and laminarin, respectively. It was specific for beta-glucans containing beta-1,3-linkages with an optimum temperature of 70 degrees C at pH 6.0. The N-terminal SLH modules were cleaved from the protein as well in Escherichia coli as in C. thermocellum, but nevertheless bound tightly to the rest of the protein. Lic16A was located on the cell surface from which it could be purified after fractionated solubilization. Its inducible production allowed C. thermocellum to grow on beta-1,3- or beta-1,3-1,4-glucan.
 
The nucleotide sequence of clone pTT26 (3786 bp), containing the gene for 1,3-beta-glucanase LamA (laminarinase) from Thermotoga neapolitana, was determined. It contains an ORF encoding a protein of 646 aa (73328 Da). The central part of the protein is homologous to the complete catalytic domain of bacterial and some eukaryotic endo-1,3-beta-D-glucanases and belongs to family 16 of glycosyl hydrolases. This domain is flanked on both sides by one copy on each side of a substrate binding domain homologue (family II). The recombinant laminarinase protein was purified from Escherichia coli host cells in two forms, a 73 kDa and a processed 52 kDa protein, both having high specific activity towards laminarin (3100 and 2600 U mg-1, respectively) and K(m) values of 2.8 and 2.2 mg ml-1, respectively. Limited activity on 1,3-1,4-beta-glucan (lichenan) was detected (90 U mg-1). Laminarin was degraded in an endoglucanase modus, yielding glucose, laminaribiose and -triose as end products. Thus LamA classifies as an endo-1,3(4)-beta-glucanase (EC 3.2.1.6). The optimum temperature of the enzymes was 95 degrees C (73 kDa) and 85 degrees C (52 kDa) at an optimum pH of 6.2. The superior thermostability of the 73 kDa enzyme is demonstrated by incubation without substrate at 100 degrees C, where 57% of the initial activity remained after 30 min (82% at 95 degrees C). Thus, LamA is the most thermostable 1,3-beta-glucanase described to date.
 
Activity of microsomal glucan synthase fractions derived from yeast and mycelia (CCH 442) 
IC,, values for inhibitors of glucan synthesis in yeast, mycelial, and permeabilized cell assays (CCH 442) 
A systematic evaluation of the in vitro (1,3)-beta-glucan synthase assay parameters was performed using microsomes prepared from Candida albicans from either yeast or mycelial phase cells. Enzyme activities of both yeast and mycelial phase microsomes depended on the presence of guanosine-5'-O-(3-thiophosphate) and either bovine serum albumin or a detergent [W-1 (polyoxyethylene ether detergent) or Brij-35 (polyoxyethylene ether, 23 lauryl ether)]. Brij-35 was included in standard assays as it was compatible with the permeabilized whole-cell assay. Microsomes derived from both the yeast and mycelial phases generally yielded similar glucan synthase activities under a range of different assay conditions. Brij-35 significantly stabilized the enzyme, yielding a half-life of 5.6 d at 4 degrees C, compared with 0.9 d without detergent. The addition of detergent during mechanical breakage of yeast cells dramatically improved glucan synthase stability and activity. Enzyme catalysis was linear for at least 75 min with 100 micrograms protein from microsomes of yeast cells grown to mid-exponential phase, with an apparent Km for UDP-glucose of 1.1 mM. The pH and temperature optima were 7.75 and 30 degrees C, respectively. Glucan synthase activity was highest in cells derived from early mid-exponential phase and declined to a basal level by stationary phase. A permeabilization-based in situ assay for glucan synthase was developed. Cells were permeabilized with 2% (v/v) solution of toluene/methanol (1:1) and assayed for glucan synthase activity using standard reaction mixtures. Reactions were linear for 30 min and were inhibited by known inhibitors of glucan synthesis.(ABSTRACT TRUNCATED AT 250 WORDS)
 
(a) Partial alignment of the S. sioyaensis enzyme with other 1,3- β -glucanases of GHF 16. Ssio , Streptomyces sioyaensis ; 
Expression of the S. sioyaensis 1,3- β -glucanase. Proteins 
Dendrogram derived from a similarity alignment of members of CBM family 6. The tree topology was evaluated and corrected according to the results obtained from a multiple sequence alignment applying likelihood (CLUSTAL W) of the GCG program. Cthe xylanase A, Clostridium thermocellum F1/YS (AF04776) ; Cthe xylanase Z, Clostridium thermocellum NCIB 10682 (M22624) ; Cste xylanase A, Clostridium stercorarium F-9 (D13325) ; Mbis cellodextranase D, Microbispora bispora (L06134) ; Ttri clotting factor G α-subunit, Tachypleus tridentatus (D16622) ; Cmix endo-1,4-glucanase B, Cellvibrio mixtus (AF003697) ; Fsuc AxeA, Fibrobacter succinogenes S85 (AF180369) ; Ssio endo-1,3-glucanase, Streptomyces sioyaensis (AF217415) ; Aaga α-agarase, Alteromonas agarilytica GJ1B (AF121273) ; Csp. XynC, Caldicellulosiruptor sp. Rt69B.1 (AF036924) ; Csp. CelA, Caldicellulosiruptor sp. Tok7B.1 (AF078737) ; Csac xynF, Caldicellulosiruptor saccharolyticus (AF005383) ; Bpol xylanase D, Bacillus polymyxa (X57094) ; Bcir α-1,6-mannanase, Bacillus circulans TN31 (B024331).
A gene encoding 1,3-beta-glucanase was isolated from Streptomyces sioyaensis based on an activity plate assay. Analysis of the deduced amino acid sequence of the gene revealed that the matured 1,3-beta-glucanase has two functional domains separated by a stretch of nine glycine residues. The N-terminal domain shares sequence similarity with bacterial endo-1,3-beta-glucanases classified in glycosyl hydrolase family 16 (GHF 16), while the C-terminal domain is a putative carbohydrate-binding module (CBM) grouped into CBM family 6. To characterize the function of each domain, both the full-length and the CBM-truncated versions of the protein were expressed in Escherichia coli and purified to homogeneity. Biochemical data suggest that the glycosyl hydrolase domain preferentially catalyses the hydrolysis of glucans with 1,3-beta linkage, and has an endolytic mode of action. Binding assay indicated that the C-terminal CBM binds to various insoluble beta-glucans (1,3-, 1,3-1,4- and 1,4- linkages) but not to xylan, a primary binding target for most members of CBM family 6. The full-length and the CBM-truncated proteins had similar specific activity (units per mol of hydrolase domain) on soluble 1,3-beta-glucan, whereas the former had much stronger specific activity on insoluble 1,3-beta-glucans, suggesting that the C-terminal CBM enhances the activity of the S. sioyaensis 1,3-beta-glucanase against insoluble substrates, presumably by increasing the frequency of encounter events between the hydrolase domain and the substrate.
 
Biomass increase of S. halifaxensis HAW-EB4 (initial OD 600 0.1 or 0.5) during incubation at 10 6C in MB 20 medium under aerobic (a) or anaerobic conditions (c-d), in the presence of TMAO (c) or nitrate (d), or in the absence of any TEA (b).
RDX oxidation of dithionite-or NADH-reduced c-type cytochrome prepared from cells grown on various TEA
Anaerobic biotransformation of RDX (89 mM) in the presence of NADH (0.75 mM) by periplasmic proteins from cells pre-grown in the presence of TMAO (a), nitrate (c) or O 2 (air) (d), or pre-incubated in the absence of any TEA (b). Reaction volumes: (a), 1 ml; (b-d), 2 ml. The protein contents for (a), (b), (c) and (d) were 0.6, 0.65, 0.7 and 0.85 mg ml "1 , respectively.
Detection of c -type cytochromes by SDS-PAGE (4–12 %, w/v, gradient gel) analysis and haem staining. Protein loading was 140–160 m g. Haem staining was conducted by stepwise incubation of the gel with TMBZ and H 2 O 2 , as described elsewhere (Thomas 
Crude cytochromes from periplasmic proteins in TMAO-grown cells [salted out with 50-65 % (NH 4 ) 2 SO 4 saturation]. (a) Absorption spectra (2.3 mg protein ml "1 ); (b) RDX biotransformation products in the presence of NADH (4.7 mg protein ml "1 ).
Shewanella halifaxensis HAW-EB4 was previously isolated for its potential to mineralize hexahydro-1,3,5-trinitro-1,3,5-triazine (RDX) from a UXO (unexploded ordnance)-contaminated marine sediment site near Halifax Harbor. The present study was undertaken to determine the effect of terminal electron acceptors (TEA) on the growth of strain HAW-EB4 and on the enzymic processes involved in RDX metabolism. The results showed that aerobic conditions were optimal for bacterial growth, but that anaerobic conditions in the presence of trimethylamine N-oxide (TMAO) or in the absence of TEA favoured RDX metabolism. RDX as a substrate neither stimulated respiratory growth nor induced its own biotransformation. Strain HAW-EB4 used periplasmic proteins to transform RDX to both nitroso [hexahydro-1-nitroso-3,5-dinitro-1,3,5-triazine (MNX), hexahydro-1,3-dinitroso-5-nitro-1,3,5-triazine (DNX), and hexahydro-1,3,5-trinitroso-1,3,5-triazine (TNX)] and ring cleavage products (such as methylenedinitramine), with more nitroso formation in cells grown on TMAO or pre-incubated in the absence of TEA. Using spectroscopy, SDS-PAGE and haem-staining analysis, strain HAW-EB4 was found to produce different sets of c-type cytochromes when grown on various TEA, with several more cytochromes produced in cells grown on TMAO. Crude cytochromes from total periplasmic proteins of TMAO-grown cells metabolized RDX to products similar to those found in assays using total periplasmic proteins and whole cells. To prove the involvement of cytochrome in RDX metabolism, we monitored dithionite- or NADH-reduced cytochromes by their absorbance at the alpha (551 nm) or gamma (418-420 nm) bands during anaerobic incubation with RDX. In both cases we found that RDX biotransformation was accompanied by oxidation of reduced cytochrome. Furthermore, O(2), an oxidant of reduced cytochrome, inhibited RDX transformation. The present results demonstrate that S. halifaxensis HAW-EB4 metabolizes RDX optimally under TMAO-reducing conditions, and that c-type cytochromes are involved.
 
Lipooligosaccharide (LOS) is a major virulence factor of the pathogenic Neisseria. Three galactosyltransferase genes, lgtB, lgtE and lgtH, responsible for the biosynthesis of LOS oligosaccharide chains, were analysed in five Neisseria species. The function of lgtH in Neisseria meningitidis 6,275 was determined by mutagenesis and chemical characterization of the parent and mutant LOS chains. The chemical characterization included SDS-PAGE, immunoblot, hexose and mass spectrometry analyses. Compared with the parent LOS, the mutant LOS lacked galactose, and its oligosaccharide decreased by three or four sugar units in matrix-assisted laser desorption ionization (MALDI)-MS analysis. The results show that lgtH encodes a beta-1,4-galactosyltransferase, and that the glucose moiety linked to heptose (Hep) in the alpha chain is the acceptor site in the biosynthesis of Neisseria LOS. To understand the sequence diversity and relationships of lgtB, lgtE and lgtH, the entire lgt-1 locus was further sequenced in three N. meningitidis strains and three commensal Neisseria strains, and compared with the previously reported lgt genes from Neisseria species. Comparison of the protein sequences of the three enzymes LgtB, LgtE and LgtH showed a conserved N-terminal region, and a highly variable C-terminal region, suggesting functional constraint for substrate and acceptor specificity, respectively. The analyses of allelic variation and evolution of 23 lgtB, 12 lgtE and 14 lgtH sequences revealed a distinct evolutionary history of these genes in Neisseria. For example, the splits graph of lgtE displayed a network evolution, indicating frequent DNA recombination, whereas splits graphs of lgtB and lgtH displayed star-tree-like evolution, indicating the accumulation of point mutations. The data presented here represent examples of the evolution and variation of prokaryotic glycosyltransferase gene families. These imply the existence of multiple enzyme isoforms for biosynthesis of a great diversity of oligosaccharides in nature.
 
(a) Diagrammatic representation of the domains encoded by the gene products of ORF1 and xynA. The position of the PCR primer pair XYNACBD4BF-XYNACBD4BR used to amplify the xynAd1/2 fragment from the genomic DNA of ' Caldibacillus cellulovorans ' for construction of pSUN18 is shown. The PCR primer pair XYNA-CBD4BF-XYNACBD4BR used to amplify the xynAd2 fragment from pSUN18 for the construction of the expression pSUN24 is also shown. sig, signal peptide ; *, stop codon ; PT, proline-threonine-rich linker peptide ; D, domain. (b) A plot of the relative mol % GjC content across the DNA sequence encoding the various domains of XynA.
Effect of temperature on the enzyme activity of the purified recombinant XynAd1/2 ($) and XynAd2 (#). Assays were performed for 15 min at the indicated temperature.
Thin-layer chromatograms of xylooligosaccharide hydrolysis products. Purified XynAd1/2 and XynAd2 were incubated with xylooligosaccharides for 3 h at 70 and 60 m C, respectively. Xylooligosaccharide controls without enzyme were also incubated for 3 h at the above temperatures. (a) Hydrolysis products liberated by XynAd1/2. (b) Hydrolysis products liberated by XynAd2. C, xylooligosaccharide without enzyme ; E, xylooligosaccharide with enzyme ; X 1 , xylose ; X 2 , xylobiose ; X 3 , xylotriose ; X 4 , xylotetraose ; X 5 , xylopentaose ; X 6 , xylohexaose. 
The nucleotide sequence of the complete xynA gene, encoding a novel multidomain xylanase XynA of 'Caldibacillus cellulovorans', was determined by genomic-walking PCR. The putative XynA comprises an N-terminal domain (D1), recently identified as a xylan-binding domain (XBD), homologous to non-catalytic thermostabilizing domains from other xylanases. D1 is followed by a xylanase catalytic domain (D2) homologous to family 10 glycosyl hydrolases. Downstream of this domain two cellulose-binding domains (CBD), D3 and D4, were found linked via proline-threonine (PT)-rich peptides. Both CBDs showed sequence similarity to family IIIb CBDs. Upstream of xynA an incomplete open reading frame was identified, encoding a putative C-terminal CBD homologous to family IIIb CBDs. Two expression plasmids encoding the N-terminal XBD plus the catalytic domain (XynAd1/2) and the xylanase catalytic domain alone (XynAd2) were constructed and the biochemical properties of the recombinant enzymes compared. The absence of the XBD resulted in a decrease in thermostability of the catalytic domain from 70 degrees C (XynAd1/2) to 60 degrees C (XynAd2). Substrate-specificity experiments and analysis of the main products released from xylan hydrolysis indicate that both recombinant enzymes act as endo-1, 4-beta-xylanases, but differ in their ability to cleave small xylooligosaccharides.
 
Ultrastructural alterations induced by 24 h DAB treatment in L. amazonensis promastigotes. In contrast to untreated control L. amazonensis promastigotes (A), which displayed the usual appearance of organelles such as mitochondrion–kinetoplast complex (k) and flagellar pocket with flagellum (f), the DAB- treated parasites showed mitochondrial alterations (B–E) such as fenestration (B, *) and kinetoplast (k) DNA disorganization (C, arrow). In advanced stages, many parasites showed complete mitochondrial destruction (D, E) and the organelle was only recognized by the presence of scarce remains of the cristae (D, arrows) and k-DNA (E, k). Myelin-like figures were observed in the cytoplasm and associated with the flagellar pocket (F, arrow). Magnification: A and D, 3⁄4 75 000; B 3⁄4 40 000; C 3⁄4 87 000; E, 3⁄4 33 500; F, 3⁄4 71 000. 
Lipid peroxidation in untreated and DAB-treated L. amazonensis. Lipid peroxidation was measured by detection of thiobarbituric acid reactive substances (TBARS). (A) Incubations with 1 or 10 mM DAB for 3 h significantly enhanced TBARS concentration, and putrescine addition returned the peroxidation to control levels. (B) DAB incubations for 24 h significantly diminished peroxidation, and putrescine addition only partially reversed this effect. Data in (A) are presented as means±SD of at least three independent experiments; data in (B) are representative of at least three independent assays. *P, 0.05; **P,0.001.
Effect of DAB on mitochondrial function in L. amazonensis . Promastigotes (5 3⁄4 10 6 ml ” 1 ) were treated with increasing DAB 
Polyamines are important regulators of growth and differentiation in a variety of cells, including parasitic protozoa. Promastigotes of Leishmania species have high levels of putrescine and spermidine and their growth can be inhibited by polyamine biosynthesis antagonists. The putrescine analogue 1,4-diamino-2-butanone (DAB) is microbicidal against Tritrichomonas foetus and Trypanosoma cruzi, so we tested its effects on Leishmania amazonensis proliferation, viability, organization, putrescine transport and synthesis as well as in vitro infectivity. DAB impaired promastigote proliferation dose-dependently (IC50 144 μM) and the parasite putrescine concentration was reduced by nearly 50 %. This analogue markedly inhibited both ornithine decarboxylase activity and [H3]putrescine uptake by promastigotes. Pre-treatment with DAB for 24 h led to compensatory enhancement of putrescine uptake, indicating an adaptive mechanism in DAB-treated parasites. Remarkably, DAB caused mitochondrial damage, assessed by transmission electron microscopy, and 3 h treatment with 1 mM DAB enhanced lipid peroxidation, whereas incubation with 10 mM DAB or for 24 h resulted in decreased peroxidation levels in the parasites. This effect was probably due to the loss of mitochondrial function, demonstrated by the diminished reduction of 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT), not observed in macrophages. Murine macrophages infected with L. amazonensis amastigotes treated with DAB had parasite loads significantly (P<0.05) lower than controls, presumably due to interference with putrescine uptake and/or synthesis. These results suggest that putrescine may be involved in leishmanial survival, possibly by maintaining the parasite's mitochondrial function. The use of analogues to interfere with polyamine/diamine synthesis and transport may shed light on its function in intracellular parasite survival and lead to identification of new targets for leishmaniasis chemotherapy.
 
Neighbour-joining tree based on nearly complete 16S rRNA gene sequences (1346 bp) of all type strains of validly described Nocardia species and of the novel bacterial isolates capable of degrading poly(trans-1,4-isoprene). Numbers at nodes indicate the level of bootstrap support based on neighbour-joining analysis of 1000 resampled datasets; only values above 50 % are given. Bar, 0.005 substitutions per nucleotide position.
Percentage mineralization during growth of bacteria on polyisoprenoides. (a) Effect of GP granule size on mineralization. Cultures (100 ml) in mineral salts medium containing 0.6 % (w/v) synthetic poly(trans-1,4-isoprene) with a defined granule size ($, 125-250; #, 250-500; ., 500-710; n, 710-1000; &, 1250-1400 mm) were inoculated with 40 mg cell fresh weight of isolate WE30. (b) Mineralization of GP by novel GP-degrading strains. Cultures (100 ml) in mineral salts medium containing 0.6 % (w/v) synthetic poly(trans-1,4-isoprene) (granule size, 250-500 mm) were inoculated with 25 mg fresh cell weight of the respective strains (%, L1b; e, N. takedensis; #, SEI2b; m, SH22a; $, SEII5a; n, WE30; X, SM1), or of G. polyisoprenivorans VH2 (&) or R. opacus PD630 (.) as negative controls. (c) Mineralization of poly(cis-1,4-isoprene) rubber by novel GPdegrading strains and by the NR-degrading strain G. polyisoprenivorans VH2. Cultures (100 ml) in mineral salts medium containing 0.6 % (w/v) synthetic poly(cis-1,4-isoprene) rubber (granule size, 250-500 mm) were inoculated with 40 mg cell fresh weight of the respective strains (%, L1b; e, N. takedensis; #, SEI2b; X, SM1; m, SH22a; $, SEII5a; n, WE30; &, G. polyisoprenivorans). All values represent the arithmetic mean of two individual cultures. The deviation from the arithmetic mean was 11.8 % for all measurements with an SD of 18.7 %.
Scanning electron micrographs of GP films after incubation with micro-organisms. GP films were incubated with cells of N. takedensis WE30 (a–c) and N. nova L1b (e–g). Films were incubated for 1 (a, e), 3 (b, f) and 5 weeks (c, g). (h) Surface of a GP film after 5 weeks incubation in the presence of G. polyisoprenivorans . (d) Surface of a GP film after 5 weeks incubation in a non-inoculated sterile control. The circle in (a) highlights a cavity which is difficult to recognize in this stage. Bars, 200 m m. 
Scanning electron micrographs of the surface of poly(cis-1,4-isoprene) particles after incubation with G. polyisoprenivorans VH2 (b), N. takedensis WE30 (c) and N. nova L1b (d) for 5 weeks. (a) Surface of poly(cis-1,4isoprene) after 5 weeks incubation under the same conditions in the absence of bacterial cells. Bars, 200 mm.
Gutta percha, the trans-isomer of polyisoprene, is being used for several technical applications due to its resistance to biological degradation. In the past, several attempts to isolate micro-organisms capable of degrading chemically pure poly(trans-1,4-isoprene) have failed. This is the first report on axenic cultures of bacteria capable of degrading gutta percha. From about 100 different habitats and enrichment cultures, six bacterial strains were isolated which utilize synthetic poly(trans-1,4-isoprene) as sole carbon and energy source for growth. All isolates were assigned to the genus Nocardia based on 16S rRNA gene sequences. Four isolates were identified as strains of Nocardia nova (L1b, SH22a, SEI2b and SEII5a), one isolate was identified as a strain of Nocardia jiangxiensis (SM1) and the other as a strain of Nocardia takedensis (WE30). In addition, the type strain of N. takedensis obtained from a culture collection (DSM 44801(T)) was shown to degrade poly(trans-1,4-isoprene). Degradation of poly(trans-1,4-isoprene) by these seven strains was verified in mineralization experiments by determining the release of CO(2). All seven strains were also capable of mineralizing poly(cis-1,4-isoprene) and to use this polyisoprenoid as a carbon and energy source for growth. Mineralization of poly(trans-1,4-isoprene) after 80 days varied from 3 % (strain SM1) to 54 % (strain SEI2b) and from 34 % (strain L1b) to 43 % (strain SH22a) for the cis-isomer after 78 days. In contrast, Gordonia polyisoprenivorans strain VH2, which was previously isolated as a potent poly(cis-1,4-isoprene)-degrading bacterium, was unable to degrade poly(trans-1,4-isoprene). Scanning electron microscopy revealed cavities in solid materials prepared from poly(trans-1,4-isoprene) and also from poly(cis-1,4-isoprene) after incubation with N. takedensis strain WE30 or with N. nova strain L1b, whereas solid poly(trans-1,4-isoprene) material remained unaffected if incubated with G. polyisoprenivorans strain VH2 or under sterile conditions.
 
Purification of CelY Activity was determined under standard assay conditions with acid-swollen Avicel as the substrate.
SDS-PAGE of the fractions obtained during purification of recombinant CelY from E. coli cell extract. Lane 1, molecular mass markers; lane 2, crude extract of E. coli XL1-blue; lane 3, crude extract of E. coli XL1-Blue(pBSce/Y); lane 4, heat-treated extract; lane 5, pooled fractions from the Q Sepharose column; lane 6, Avicelase II purified from C. stercorariurn.  
The nucleotide sequence of the celY gene coding for the thermostable exo-1,4-beta-glucanase Avicelase II of Clostridium stercorarium was determined. The gene consists of an ORF of 274Z bp which encodes a preprotein of 914 amino acids with a molecular mass of 103 kDa. The signal-peptide cleavage site was identified by comparison with the N-terminal amino acid sequence of Avicelase II purified from C stercorarium. The celY gene is located in close vicinity to the celZ gene coding for the endo-1,4-beta-glucanase Avicelase I. The CelY-encoding sequence was isolated from genomic DNA of C. stercorarium with the PCR technique. The recombinant enzyme produced in Escherichia coli as a LacZ'-CelY fusion protein could be purified using a simple two-step procedure. The properties of CelY proved to be consistent with those of Avicelase II purified from C. stercorarium. Sequence comparison revealed that CelY consists of an N-terminal catalytic domain flanked by a domain of 95 amino acids with unknown function joined to a type III cellulose-binding domain. The catalytic domain belongs to the recently proposed family L of cellulases (family 48 of glycosyl hydrolases).
 
Chitinase activity and chitin-binding capacity of truncated ChiA proteins. (a) Glycol chitin gel. Dark area indicates hydrolysis of chitin after SDS-PAGE. (b) SDS-PAGE of chitin-binding proteins. Protein extracts were added to chitin, then bound proteins were eluted and analysed by SDS-PAGE. Lanes: K, vector alone [pBluescript II KS(-)I; 1, pChiAl; 2, pChiA2 (missing presumptive chitin-binding domain).  
To examine the ecology and evolution of microbial chitinases, especially the chitin-binding domain, one of the chitinase genes (chiA) from the marine bacterium Vibrio harveyi was analysed. The deduced amino acid sequence of ChiA is not very similar overall to other proteins, except for two regions, the putative catalytic and chitin-binding domains. Among all bacterial chitinases sequenced to date, there is no relationship between percentage similarity of catalytic domains and chitin-binding domains in pairwise comparisons, suggesting that these two domains have evolved separately. The chitin-binding domain appears to be evolutionarily conserved among many bacterial chitinases and is also somewhat similar to cellulose-binding domains found in microbial cellulases and xylanases. To investigate the role of the chitin-binding domain, clones producing versions of ChiA with or without this domain were examined. One version with the domain (ChiA1) bound to and hydrolysed chitin, whereas a truncated ChiA without the putative chitin-binding domain (ChiA2) did not bind to chitin, but it could hydrolyse chitin, although not as well. ChiA1 diffused more slowly in agarose containing colloidal chitin than ChiA2, but diffusion of the two proteins in agarose without colloidal chitin was similar. These results indicate that the chitin-binding domain helps determine the movement of chitinase along N-acetylglucosamine strands and within environments containing chitin.
 
The metabolism of inositol 1,4,5-trisphosphate [Ins(1,4,5)P3] was examined in yeast cells and germ tubes of Candida albicans. Methods have been developed for analysis of the two key metabolic enzymes, Ins(1,4,5)P3, kinase and phosphatase. ATP-dependent Ins(1,4,5)P3 kinase activity was detected predominantly in the soluble fraction of cell extracts and exhibited a Km of approximately 9 microM. The apparent Km of Ins(1,4,5)P3 phosphatase for Ins(1,4,5)P3 was approximately 480 microM. The slow rate of dephosphorylation of Ins(1,4,5_P3 to inositol bisphosphate suggests a lower importance of the phosphatase within cells compared to the kinase. Since both yeast cells and germ tubes of C. albicans rapidly phosphorylated Ins(1,4,5)P3 to inositol tetrakisphosphate and inositol penta/hexakisphosphate, it is suggested that Ins(1,4,5)P3 has an important role as a precursor for production of these compounds. A sustained increase in cellular Ins(1,4,5)P3 levels was observed during germ tube formation and, prior to the onset of germination between 1 and 2 incubation, the Ins(1,4,5)P3 content increased up to eightfold. Transient increases in the level of Ins(1,4,5)P3 were also observed during yeast-like growth of C. albicans. The possible role and relative importance of Ins(1,4,5)P3 as a precursor for inositol polyphosphates and in signal transduction involving Ca2+ release from internal stores is discussed.
 
Binding of CbbR1 and CbbRm to the promoter region of the cbb genes. (a) Gel shift assays using CbbR1 with DNA fragments containing the promoter regions of cbbLS-1 , cbbLS-2 , cbbM and cbbRm . The labelled fragments were incubated with purified CbbR1. The protein concentrations used for the cbbLS-2 , cbbM and cbbRm promoter regions were as follows (in ng per 20 m l reaction mixture): 0 (lane 1), 10 (lane 2), 50 (lane 3), 250 (lane 4). For the cbbLS-1 promoter region, the protein concentrations were as follows (in ng per 20 m l reaction mixture): 0 (lane 1), 5 (lane 2), 10 (lane 3), 20 (lanes 4, 7–9), 50 (lane 5), 100 (lane 6). A 50-fold molar excess of unlabelled fragment of the cbbLS-1 (lane 7), cbbLS-2 (lane 8) or cbbM (lane 9) promoter was added to the mixture prior to the addition of labelled fragments. (b) Gel shift assays using CbbRm with DNA fragments containing the promoter regions of cbbLS-1 , cbbLS-2 , cbbM and cbbRm . The labelled fragment was incubated with purified CbbRm. The protein concentrations used for cbbLS-1 , cbbLS-2 and cbbRm promoter regions were as follows (in ng per 20 m l reaction mixture): 0 (lane 1), 10 (lane 2), 50 (lane 3), 250 (lane 4). For the cbbM promoter region, the protein concentrations were as follows (in ng per 20 m l reaction mixture): 0 (lane 1), 5 (lane 2), 10 (lane 3), 20 (lanes 4, 7–9), 50 (lane 5), 100 (lane 6). A 50-fold molar excess of unlabelled fragment of the cbbLS-1 (lane 7), cbbLS-2 (lane 8) or cbbM (lane 9) promoter was added to the mixture prior to the addition of labelled fragments. The open and filled arrow boxes indicate free DNA probe and DNA–protein complex, respectively. Asterisks indicate contaminated non-specific bands. 
Western immunoblots of (a) H. marinus wild-type strain MH-110, (b) cbbR1 strain dR1, (c) cbbRm strain dRm and (d) cbbR1 cbbRm double mutant strain ddR using antibodies for CbbS-1 (left), CbbS-2 (centre) and CbbM (right). Cell-free extracts (approx. 10 mg total protein) of H. marinus strains grown at 15 % (lane 1), 2 % (lane 2), 0?15 % (lane 3) and 0?03 % (lane 4) CO 2 concentrations were used. The samples from the wild-type strain and each mutant were analysed on the same gel to allow a quantitative comparison. Panel (a) shows a representative result from each analysis. 
Growth profiles of the wild-type and the cbbR1 cbbRm mutant ddR. The wild-type and strain ddR were cultivated under four CO 2 conditions (percentage CO 2 concentrations shown). $, Wild-type; ., ddR. The data are representative of duplicate experiments. 
Ribonuclease protection analyses of the cbbR1 and cbbRm transcripts in wild-type strain MH-110 under different CO 2 conditions. (a) The cbbR1 transcript was observed using a 411 bp riboprobe containing the last 107 codons of cbbR1. The expected size of the RNase-protected fragments of this cbbR1 transcript was 321 bp. (b) The cbbRm transcript was detected with a 440 bp riboprobe containing the last 110 codons of cbbRm. The expected size of the RNase-protected fragments of this transcript was 330 bp. The cbbR1 or cbbRm probe was hybridized to yeast RNA in the absence (lane 1) or presence (lane 2) of added RNase A/T1. The probes were hybridized with RNA preparations from cells grown at 15 % (lane 3), 2 % (lane 4), 0?15 % (lane 5) or 0?03 % (lane 6) CO 2. Ten (a) and 20 mg (b) total RNA were used to detect the cbbR1 and cbbRm transcripts, respectively. The open and filled arrowheads indicate the size of undigested probe and a protected band, respectively. Results are representative of three independent experiments with similar results. 
Hydrogenovibrio marinus MH-110 possesses three different sets of genes for ribulose 1,5-bisphosphate carboxylase/oxygenase (RubisCO): two form I (cbbLS-1 and cbbLS-2) and one form II (cbbM). We have previously shown that the expression of these RubisCO genes is dependent on the ambient CO2 concentration. LysR-type transcriptional regulators, designated CbbR1 and CbbRm, are encoded upstream of the cbbLS-1 and cbbM genes, respectively. In this study, we revealed by gel shift assay that CbbR1 and CbbRm bind with higher affinity to the promoter regions of cbbLS-1 and cbbM, respectively, and with lower affinity to the other RubisCO gene promoters. The expression patterns of the three RubisCOs in the cbbR1 and the cbbRm gene mutants showed that CbbR1 and CbbRm were required to activate the expression of cbbLS-1 and cbbM, respectively, and that neither CbbR1 nor CbbRm was required for the expression of cbbLS-2. The expression of cbbLS-1 was significantly enhanced under high-CO2 conditions in the cbbRm mutant, in which the expression of cbbM was decreased. Although cbbLS-2 was not expressed under high-CO2 conditions in the wild-type strain or the single cbbR mutants, the expression of cbbLS-2 was observed in the cbbR1 cbbRm double mutant, in which the expression of both cbbLS-1 and cbbM was decreased. These results indicate that there is an interactive regulation among the three RubisCO genes.
 
Nucleotide sequence identity of RubisCO genes from strain 5185-9A1 with those from other organisms 
The Gram-negative bacterium strain S185-9A1 is a novel marine alpha-proteobacterium that oxidizes manganese (II) to manganese (IV). Initial DNA hybridization screening showed that S185-9A1 possesses a gene similar to cbbL, the gene coding for the large subunit of ribulose-1,5-bisphosphate carboxylase/oxygenase (Rubis CO; EC 4.1.1.39). However, no RubisCO enzyme activity was found in cultures of S185-9A1. Genes coding for both large (cbbL) and small (cbbS) subunits of a RubisCO enzyme were identified, isolated and sequenced. When these genes were introduced into an Escherichia coli host strain, ribulose-1,5-bisphosphate-dependent CO2 fixation occurred under control of a lac promoter, indicating that the protein is functional in E. coli. Although their function is unknown, this is the first direct evidence for the presence of RubisCO genes in a manganese-oxidizing bacterium. Phylogenetic analysis of the RubisCO genes of strain S185-9A1 showed that they are divergent, but are more related to those from non-chlorophyte algal chloroplasts than are those from other bacteria. The fact that the RubisCO sequence of strain S185-9A1 is not closely related to any other published RubisCO sequence suggests that the protein may be valuable for studies of the function and evolution of the RubisCO enzyme as well as its activity in the environment.
 
Analysis of the nucleotide sequence of the form I ribulose-1,5-bisphosphate carboxylase/oxygenase (RubisCO) genes (cbbL and cbbS) of the non-sulfur purple bacterium Rhodobacter capsulatus indicated that the deduced amino acid sequence of the large subunit was not closely homologous to the large subunit from related organisms. Indeed, phylogenetic analysis suggested that the large subunit protein (CbbL) more closely resembled the enzyme from alpha/beta/gamma purple bacteria and cyanobacteria and is within a 'green-like' radiation of the RubisCO phylogenetic tree, well separated from CbbL of the related organism Rhodobacter sphaeroides. A cbbQ gene was discovered downstream of cbbS in Rh. capsulatus, a gene arrangement which also appears to be limited to certain organisms containing a 'green-like' RubisCO. Upstream, and divergently transcribed from cbbLSQ, is a gene (cbbRI) that encodes a LysR-type transcriptional activator. Phylogenetic analysis of the deduced amino acid sequence of CbbRI also suggests that this protein is quite distinct from the Rh. sphaeroides CbbR protein, and is even distinct from the previously described CbbRII protein, the gene of which is upstream and divergently transcribed from the cbbII operon of Rh. capsulatus. Interestingly, Rh. capsulatus CbbRI is more closely related to CbbR from bacteria whose RubisCO falls within the 'green-like' radiation of the CbbL tree. These studies suggest that the cbbRI-cbbL-cbbS-cbbQ genes were acquired by Rh. capsulatus via horizontal gene transfer from a bacterial species containing a 'green-like' RubisCO.
 
The occurrence of the different genes encoding ribulose-1,5-bisphosphate carboxylase/oxygenase (RubisCO), the key enzyme of the Calvin-Benson-Bassham cycle of autotrophic CO(2) fixation, was investigated in the members of the genus Thiomicrospira and the relative genus Thioalkalimicrobium, all obligately chemolithoautotrophic sulfur-oxidizing Gammaproteobacteria. The cbbL gene encoding the 'green-like' form I RubisCO large subunit was found in all analysed species, while the cbbM gene encoding form II RubisCO was present only in Thiomicrospira species. Furthermore, species belonging to the Thiomicrospira crunogena 16S rRNA-based phylogenetic cluster also possessed two genes of green-like form I RubisCO, cbbL-1 and cbbL-2. Both 16S-rRNA- and cbbL-based phylogenies of the Thiomicrospira-Thioalkalimicrobium-Hydrogenovibrio group were congruent, thus supporting its monophyletic origin. On the other hand, it also supports the necessity for taxonomy reorganization of this group into a new family with four genera.
 
The presence and diversity of the cbb genes encoding the large subunit of ribulose-1,5-bisphosphate carboxylase/oxygenase (RuBisCO) (a key enzyme of the Calvin-Benson cycle of autotrophic CO(2) assimilation) were investigated in pure cultures of seven genera of halophilic chemolithoautotrophic sulfur-oxidizing bacteria (SOB) and in sediments from a hypersaline lake in which such bacteria have been recently discovered. All of the halophilic SOB strains (with the exception of Thiohalomonas nitratireducens) possessed the cbbL gene encoding RuBisCO form I, while the cbbM gene encoding RuBisCO form II was detected only in some of the pure cultures. The general topologies of the CbbL/CbbM trees and the 16S rRNA gene tree were different, but both markers showed that the halophilic SOB genera formed independent lineages in the Gammaproteobacteria. In some cases, such as with several strains of the genus Thiohalospira and with Thioalkalibacter halophilus, the cbbL clustering was incongruent with the positions of these strains on the ribosomal tree. In the cbbM tree, the clustering of Thiohalospira and Thiohalorhabdus strains was incongruent with their branching in both cbbL and 16S rRNA gene trees. cbbL and cbbM genes related to those found in the analysed halophilic SOB were also detected in a sediment from a hypersaline lake in Kulunda Steppe (Russia). Most of the cbbL and cbbM genes belonged to members of the genus Thiohalorhabdus. In the cbbL clone library, sequences related to those of Halothiobacillus and Thiohalospira were detected as minor components. Some of the environmental cbbM sequences belonged to as yet unknown phylotypes, representing deep lineages of halophilic autotrophs.
 
Phylogenetic tree based on the complete nucleotide sequences of the STSym (16S)-1 OTU and other species retrieved from the DDBJ database. The tree was constructed using the NJ method with the Kimura-2 correction parameter. Bootstrap values, calculated from 1000 replicates, are expressed as percentages ; only those values greater than 50 % are shown at the nodes of the tree. α, β, ε, and γ represent the different classes of the Proteobacteria. Bar, 0n1 substitutions per site.
Phylogenetic tree based on the deduced partial RuBisCO cbbM amino acid sequences of ST-Sym(II)-1 and ST-Sym(II)-2 and those retrieved from the databases. The tree was constructed using the NJ method with the Kimura-2 correction parameter. Bootstrap values, calculated from 1000 replications, are expressed as percentages and only those values greater than 50 % are shown at the nodes of the tree. The Greek
Alignments of the 16S rDNA sequences in the target region of the Lamellibrachia sp. symbiont-specific probe Lam384R [ST-Sym(16S)-1 sequence]. The shaded regions represent the mismatches with the corresponding nucleotides in the STSym(16S)-1 sequence. The accession numbers for the species shown in the alignment are : ST-Sym(16S)-1, AB042416 ; Lamellibrachia sp. endosymbiont, D83060 ; Lamellibrachia columna endosymbiont, U77481 ; Riftia pachyptila endosymbiont, U77478 ; Rhodobacter sulfidophilus, D13475 ; Rhodospirillum rubrum, D30778 ; Bathymodiolus endosymbiont, AB056868 ; E. coli, J01859. The letters in parentheses (α, ε, γ) represent the classification of the species within the Proteobacteria. The numbers of the aligned 16S rDNA positions are shown at the right side.
For legend see facing page. 
To better understand the contribution of micro-organisms to the primary production in the deep-sea gutless tubeworm Lamellibrachia sp., the 16S-rDNA-based phylogenetic data would be complemented by knowledge of the genes that encode the enzymes relevant to chemoautotrophic carbon fixation, such as D-ribulose 1,5-bisphosphate carboxylase-oxygenase (RuBisCO; EC 4.1.1.39). To phylogenetically characterize the autotrophic endosymbiosis within the trophosome of the tubeworm Lamellibrachia sp., bulk trophosomal DNA was extracted and analysed based on the 16S-rRNA- and RuBisCO-encoding genes. The 16S-rRNA- and RuBisCO-encoding genes were amplified by PCR, cloned and sequenced. For the 16S rDNA, a total of 50 clones were randomly selected and analysed directly by sequencing. Only one operational taxonomic unit resulted from the 16S rDNA sequence analysis. This may indicate the occurrence of one endosymbiotic bacterial species within the trophosome of the Lamellibrachia sp. used in this study. Phylogenetic analysis of the 16S rDNA showed that the Lamellibrachia sp. endosymbiont was closely related to the genus Rhodobacter, a member of the alpha-Protebacteria. For the RuBisCO genes, only the form II gene (cbbM) was amplified by PCR. A total of 50 cbbM clones were sequenced, and these were grouped into two operational RuBisCO units (ORUs) based on their deduced amino acid sequences. The cbbM ORUs showed high amino acid identities with those recorded from the ambient sediment bacteria. To confirm the results of sequence analysis, the localization of the symbiont-specific 16S rRNA and cbbM sequences in the Lamellibrachia sp. trophosome was visualized by in situ hybridization (ISH), using specific probes. Two types of cells, coccoid and filamentous, were observed at the peripheries of the trophosome lobules. Both the symbiont-specific 16S rDNA and cbbM probes hybridized at the same sites coincident with the location of the coccoid cells, whereas the filamentous cells showed no cbbM-specific signals. The RuBisCO form I gene (cbbL) was neither amplified by PCR nor detected by ISH. This is the first demonstration of chemoautotrophic symbiosis in the deep-sea gutless tubeworm, based on sequence data and in situ localization of both the 16S-rRNA- and RuBisCO-encoding genes.
 
Lipid utilization by M. tuberculosis leading to lipoglycan synthesis via gluconeogenesis. The genes glpX and pckA which code for enzymes catalysing irreversible steps are shown, as well as fum , which lies next to glpX in the M. tuberculosis genome. The unidirectional arrows are drawn to emphasize the directions of reactions in gluconeogenesis, and do not indicate whether reactions are inherently uni- or bidirectional. The grey arrows indicate steps in the TCA cycle that would be used with glucose, but not with fatty acids, as a carbon source. 
A multiple sequence alignment of actinomycete class II FBPases and E. coli GlpX. The putative class II FBPases for M. tuberculosis (Mtb), M. smegmatis (Msm), M. avium (Mav), M. leprae (Mlp), Streptomyces coelicolor (Sco), Corynebacterium diphtheriae (Cdi), and E. coli (Eco) GlpX and C. glutamicum (Cgl) Fbp were aligned with CLUSTALW (Thompson et al ., 1994) using default settings. The CLUSTALW alignment was shaded using the CHROMA program (Goodstadt 
Comparative genome analysis of actinomycete glpX genes. The genetic structure of regions containing the glpX gene are shown for M. tuberculosis (Mtub), M. smegmatis (Msm), M. avium (Mav), M. leprae (Mlep), Corynebacterium glutamicum (Cglu), C. diphtheriae (Cdip) and S. coelicolor (Sco). Black arrows, glpX gene; white arrows, genes showing homology to those in the M. tuberculosis phoH2-Rv1101c region; grey arrows, genes not showing homology to those in the M. tuberculosis phoH2-Rv1101c region; white hatched arrow (ML1948), as for grey arrows but pseudogene. Gene names/ numbers arrows refer to the gene identification for that species except for M. avium, which has not yet been annotated, and M. tuberculosis designations are used. Numbering of genes refers to the relevant species: thus 1100c is short for Rv1100c, 1945 is short for ML1945.
There are now abundant data indicating that Mycobacterium tuberculosis uses fatty acids as a carbon source in vivo. A key enzyme in gluconeogenesis, missing in the original annotation of the M. tuberculosis genome, is fructose 1,6-bisphosphatase (FBPase; EC 3.1.3.11). The authors have shown that M. tuberculosis Rv1099c, a glpX homologue, can complement Escherichia coli mutants lacking FBPase. The protein encoded by Rv1099c was shown to have FBPase activity. Rv1099c was expressed at significant levels in M. tuberculosis, and may encode the major FBPase of this pathogen.
 
The thermotolerant Gram-positive methylotroph Bacillus methanolicus is able to grow with methanol, glucose or mannitol as a sole carbon and energy source. Fructose 1,6-biphosphate aldolase (FBA), a key enzyme of glycolysis and gluconeogenesis, is encoded in the genome of B. methanolicus by two putative fba genes, the chromosomally located fbaC and fbaP on the naturally occurring plasmid pBM19. Their amino acid sequences share 75% identity and suggest a classification as class II aldolases. Both enzymes were purified from recombinant E. coli and were found to be active as homotetramers. Both enzymes were activated by either manganese or cobalt ions and inhibited by ADP, ATP and EDTA. The kinetic parameters allowed to distinguish the chromosomally encoded FBAC from the plasmid encoded FBAP since FBAC showed higher affinity towards fructose 1,6-bisphosphate (KM of 0.16 ± 0.01 mM as compared to 2 ± 0.08 mM) as well as higher glycolytic catalytic efficiency (31.3 as compared to 0.8 s-1 mM-1) than FBAP. On the other hand, FBAP exhibited a higher catalytic efficiency in gluconeogenesis (50.4 as compared to 1.4 s-1 mM-1 with dihydroxyacetone phosphate and 4 as compared to 0.4 s-1 mM-1 with GAP as limiting substrate). The aldolase-negative C. glutamicum mutant Δfda could be complemented with both FBA genes from B. methanolicus. Based on the kinetic data, we propose that FBAC acts as major aldolase in glycolysis whereas FBAP acts as major aldolase in gluconeogenesis in B. methanolicus.
 
Lipophilic yeasts of the genus Malassezia are associated with several skin diseases, such as pityriasis versicolor, Malassezia folliculitis, seborrhoeic dermatitis and atopic dermatitis, and are also increasingly associated with catheter-related fungaemia. The cell wall components of pathogenic micro-organisms behave as an antigen and/or ligand of the innate immune response. Live cells of Malassezia furfur and Malassezia pachydermatis did not react with an anti-alpha-1,2-mannoside antibody. However, they showed a strong hydrophobicity and reactivity with an anti-beta-1,3-glucan antibody compared to those of C. albicans. The cell wall polysaccharides of M. furfur and M. pachydermatis were isolated and their structures analysed by (1)H and (13)C NMR experiments. Both polysaccharides were shown to be beta-1,6-linked linear galactofuranosyl polymers with a small amount of mannan. The presence of galactomannan on cells of Malassezia species has not been described previously. The galactomannan did not react with an anti-Aspergillus fumigatus monoclonal antibody which has specificity for beta-1,5-linked galactofuranosyl residues. An anti-M. furfur antibody strongly reacted with the galactomannans of M. furfur and M. pachydermatis, but did not react with the galactomannans of Trichophyton rubrum, A. fumigatus or Fonsecaea pedrosoi. The characteristics of the anti-M. furfur antibody suggest that there is potential for diagnosis of Malassezia infections by antigen detection.
 
Although ROT1 is essential for growth of Saccharomyces cerevisiae strain BY4741, the growth of a rot1Δ haploid was partially restored by the addition 0.6 M sorbitol to the growth medium. Rot1p is predicted to contain 256 amino acids, to have a molecular mass of 29 kDa, and to possess a transmembrane domain near its C-terminus. Candida albicans and Schizosaccharomyces pombe have Rot1p homologues with high identity that also have predicted transmembrane domains. To explore the role of Rot1p, the phenotypes of the rot1Δ haploid were analysed. Deletion of ROT1 caused cell aggregation and an abnormal morphology. Analysis of the cell cycle showed that rot1Δ cells are delayed at the G2/M phase. The rot1Δ cells were resistant to K1 killer toxin and hypersensitive to SDS and hygromycin B, suggesting that they had cell wall defects. Indeed, greatly reduced levels of alkali-soluble and -insoluble 1,6-β-glucan, and increased levels of chitin and 1,3-β-glucan, were found in rot1Δ cells. Furthermore, the phenotypes of rot1Δ cells resemble those of disruption mutants of the KRE5 and BIG1 genes, which show greatly reduced levels of cell wall 1,6-β-glucan. Incorporation of glycosylphosphatidylinositol (GPI -dependent cell wall proteins in big1Δ and rot1Δ cells was examined using a (GFP-Flo1 fusion protein. GFP fluorescence was detected both on the cell surface and in the culture medium, suggesting that, in these mutants, mannoproteins may become only weakly bound to the cell wall and some of these proteins are released into the medium. Electron microscopic analyses of rot1Δ and big1Δ cells showed that the electron-dense mannoprotein rim staining was more diffuse and paler than that in the wild-type, and that the outer boundary of the cell wall was irregular. A big1Δrot1Δ double mutant had a growth rate similar to the corresponding single mutants, suggesting that Rot1p and Big1p have related functions in 1,6-β-glucan synthesis.
 
The use of a novel monoclonal antibody (mAb) that reacts with (1,6)-beta-glucan has permitted the study of the different covalent linkages between glucan and mannoproteins in the cell wall of Candida albicans. The mAb JRR1 was originally raised by immunization with Zymolyase extracts from C. albicans cell walls, but it soon became apparent that it reacted with a (1,6)-beta-glucan epitope. By using this antibody, we show the existence of glucan-mannoprotein complexes between the (1,6)-beta-glucan epitope recognized by the antibody and cell wall mannoproteins. The topology of the (1,6)-beta-glucan in the cell wall of C. albicans has also been studied.
 
KIC1 encodes a PAK kinase that is involved in morphogenesis and cell integrity. Both over- and underexpressing conditions of KIC1 affected cell wall composition. Kic1-deficient cells were hypersensitive to the cell wall perturbing agent calcofluor white and had less 1,6-beta-glucan. When Kic1-deficient cells were crossed with various kre mutants, which also have less 1,6-beta-glucan in their wall, the double mutants displayed synthetic growth defects. However, when crossed with the 1,3-beta-glucan-deficient strain fks1delta, no synthetic growth defect was observed, supporting a specific role for KIC1 in regulating 1,6-beta-glucan levels. Kic1-deficient cells also became highly resistant to the cell wall-degrading enzyme mixture Zymolyase, and exhibited higher transcript levels of the cell wall protein-encoding genes CWP2 and SED1. Conversely, overexpression of KIC1 resulted in increased sensitivity to Zymolyase and in a higher level of 1,6-beta-glucan. Multicopy suppressor analysis of a Kic1-deficient strain identified RHO3. Consistent with this, expression levels of RHO3 correlated with 1,6-beta-glucan levels in the cell wall. Interestingly, expression levels of KIC1 and the MAP kinase kinase PBS2 had opposite effects on Zymolyase sensitivity of the cells and on cell wall 1,6-beta-glucan levels in the wall. It is proposed that Kic1 affects cell wall construction in multiple ways and in particular in regulating 1,6-beta-glucan levels in the wall.
 
Glycolytic and gluconeogenic pathway in B. phymatum. EC 1.1.1.44, 6-phosphogluconate dehydrogenase; EC 1.2.1.12, glyceraldehyde-3-phosphate dehydrogenase; EC 2.7.1.2, glucokinase; EC 2.7.1.4, fructokinase; EC 2.7.1.11, 6- phosphofructokinase; EC 2.7.1.12, gluconokinase; EC 2.7.1.40, pyruvate kinase; EC 2.7.2.3, phosphoglycerate kinase; EC 3.1.3.11, FBPase; EC 4.1.2.13, fructose 1,6-bisphosphate aldolase; EC 4.1.2.14, 2-dehydro-3-deoxyphosphogluconate aldolase; EC 4.2.1.11, enolase; EC 4.2.1.12, phosphogluconate dehydratase; EC 5.3.1.9, glucose-6-phosphate isomerase; EC 5.4.2.1, phosphoglycerate mutase. 
Bacterial strains and plasmids used in this study
Nodulation test of the parental strain and Tn 5 -induced mutants of B. phymatum. (a) Nodules induced by B. phymatum STM815 (left) and the mutant strain KM16-22 (right) at 28 days after inoculation on the roots of M. pudica ; (b) wild-type STM815 can form root nodules on M. pudica ; (c) mutant KM16-22 and (d) mutant KM51 have no ability to form root nodules, but they still cause root hair deformation; (e) M. pudica root without bacterial inoculation. Bars, 1 mm. 
Growth of B. phymatum STM815, KM16-22 and KM51 on complex and minimal media containing various carbon sources ND, Not determined.
Relative enzyme activity and nodulation response of the parental strain, mutant strain and the mutant strain harbouring the uninterrupted enzyme gene
Burkholderia phymatum STM815 is a β-rhizobial strain that can effectively nodulate several species of the large legume genus Mimosa. Two Tn5-induced mutants of this strain, KM16-22 and KM51, failed to form root nodules on Mimosa pudica, but still caused root hair deformation, which is one of the early steps of rhizobial infection. Both mutants grew well in a complex medium. However, KM16-22 could not grow on minimal medium unless a sugar and a metabolic intermediate such as pyruvate were provided, and KM51 also could not grow on minimal medium unless a sugar was added. The Tn5-interrupted genes of the mutants showed strong homologies to pgm, which encodes 2,3-biphosphoglycerate-dependent phosphoglycerate mutase (dPGM), and fbp, which encodes fructose 1,6-bisphosphatase (FBPase). Both enzymes are known to be involved in obligate steps in carbohydrate metabolism. Enzyme assays confirmed that KM16-22 and KM51 had indeed lost dPGM and FBPase activity, respectively, whilst the activities of these enzymes were expressed normally in both free-living bacteria and symbiotic bacteroids of the parental strain STM815. Both mutants recovered their enzyme activity after the introduction of wild-type pgm or fbp genes, were subsequently able to use carbohydrate as a carbon source, and were able to form root nodules on M. pudica and to fix nitrogen as efficiently as the parental strain. We conclude that the enzymes dPGM and FBPase are essential for the formation of a symbiosis with the host plant.
 
In this study, the identification and characterization of the Yarrowia lipolytica homologues of Saccharomyces cerevisiae alpha-1,6-mannosyltransferases Anp1p and Och1p, designated YlAnl1p and YlOch1p, are described. In order to confirm the function of the Y. lipolytica proteins, including the previously isolated YlMnn9p, in the N-glycosylation pathway, a phenotypic analysis of the disrupted strains Delta Ylmnn9, Delta Ylanl1, Delta Yloch1, Delta Ylanl1 Delta Ylmnn9 and Delta Ylmnn9 Delta Yloch1 was performed. Disruption of the YlMNN9, YlANL1 and YlOCH1 genes caused an increased sensitivity to SDS, compatible with a glycosylation defect, and to Calcofluor White, characteristic of cell-wall defects. Moreover, Western-blot analysis of a heterologous glycosylated protein confirmed a direct role of YlMnn9p and YlAnl1p in the N-glycosylation process. These mutant strains, Delta Ylmnn9, Delta Ylanl1, Delta Yloch1, Delta Ylanl1 Delta Ylmnn9 and Delta Ylmnn9 Delta Yloch1 may thus be used to establish a model for the Y. lipolytica N-linked glycosylation pathway.
 
Properties of the four catechol meta cleavage enzymes observed under the experimental conditions used 
Purification of 3-sulphocatechol 2,3-dioxygenase 
SDS-PAGE of purified 3SC230. Proteins were stained with Coomassie blue in 12% gels. Track A, 3SC230 (3 pg protein); track B, standard protein markers. 
Alcaligenes sp. strain O-1 utilizes three sulphonated aromatic compounds as sole sources of carbon and energy for growth in minimal salts medium-benzenesulphonate (BS), 4-toluenesulphonate (TS) and 2-aminobenzenesulphonate (2AS). The degradative pathway(s) in 2AS-grown cells are initiated with membrane transport, NADH-dependent dioxygenation and meta ring cleavage. The specific activity of the NADH-dependent dioxygenation(s) varied with the growth phase and was maximal near the end of exponential growth for each growth substrate. Cells were harvested at this point from BS-, TS- and 2AS-salts medium. Cells grown with each sulphonated substrate could oxygenate all three compounds, but only 2AS-grown cells consumed 2 mol O2 per mol 2AS or BS or TS. BS- and TS-grown cells consumed 2 mol O2 per mol BS or TS but failed to oxygenate the product of oxygenation of 2AS, 3-sulphocatechol (3SC). These observations were repeated with cell extracts and we concluded that there were two sets of desulphonative pathways in the organism, one for 2AS and one for BS and TS. We confirmed this hypothesis by separating the degradative enzymes from 2AS-, BS- or TS-grown cells. A 2AS dioxygenase system and a 3SC-2,3-dioxygenase (3SC23O) were detected in 2AS-grown cells only. In both BS- and TS-grown cells a dioxygenase system for BS and TS was observed as well as a principal catechol 2,3-dioxygenase (C23O-III), neither of which was present in 2AS-grown cells. The 3SC23O was purified to near homogeneity, found to be monomeric (M(r) 42,000), and to catalyse 2,3-dioxygenation to a product that decayed spontaneously to sulphite and 2-hydroxymuconate. The 2AS dioxygenase system could cause not only deamination of 2AS but also desulphonation of BS and TS. The BS dioxygenase could desulphonate BS and apparently either desulphonate or deaminate 2AS. Strain O-1 thus seems to contain two putative, independently regulated operons involving oxygenation and spontaneous desulphonation(s). One operon encodes at least the 2AS dioxygenase system and 3SC23O whereas the other encodes at least the BS/TS dioxygenase system and C23O-III.
 
A (1-->3)-beta-D-glucan glucanohydrolase (EC 3.2.1.6), capable of hydrolysing resistant curdlan, was purified chromatographically from the culture supernatant of Bacillus circulans complex YK9 on Toyopearl HW-55F and butyl-Toyopearl 650M columns. The purified enzyme had a specific activity of 190 units mg-1 on regenerated curdlan. The molecular mass was estimated to be about 70 kDa as judged by SDS-PAGE. The enzyme had a pH optimum of approximately pH 6.0. It hydrolysed regenerated and resistant curdlans yielding predominantly laminari-biose, although the rate of hydrolysis of the former was much higher than the latter. This enzyme rapidly hydrolysed laminaran, curdlan and carboxymethyl-curdlan, but did not cleave schizophyllan and screloglucan, which have glucosyl side chains. The enzyme hydrolysed low molecular mass (1-->3)-beta-D-glucans-(mean degree of polymerization, DPn = 131, 49 and 14) and laminari-heptaose more efficiently than curdlan. It also hydrolysed laminari-hexaose and -pentaose effectively, but laminari-tetraose only slightly and it did not hydrolyse laminari-triose or -biose. The enzyme is an exo-hydrolase of curdlan and various oligomers composed of (1-->3)-beta-D-glucosidic linkages, liberating laminari-biose from their non-reducing terminals. The laminari-biose generated was in the alpha-form.
 
The topological distribution of two epitopes in the cell wall of Candida albicans, the kinetics of their incorporation into the regenerating protoplast wall, and the effect of different antibiotics upon their incorporation and localization have been studied. To do so, two monoclonal antibodies that react against an O-glycosylated mannoprotein (1B12) and against a 1,6-beta-glucan epitope (JRR1) were used. The results show that the JRR1 epitope is localized in an internal layer of the cell wall, in contrast to the 1B12 epitope, which is superficial, and that the incorporation of the JRR1 epitope into walls of regenerating protoplasts precedes that of the 1B12 epitope. The JRR1 epitope is normally found in the culture medium of control cells, but not in that of papulacandin-B-treated cells, and tunicamycin interferes with the incorporation of the 1B12 epitope into the cell walls. Finally, the results support the hypothesis that mannoproteins are not 1,6-beta-glycosylated before their secretion.
 
The approximately 10 kbp region encompassing nprB and argJ at 102 degrees on the Bacillus subtilis chromosome was sequenced, revealing 12 ORFs, four known genes (argJ, argC, ipi and nprB) and two genes, yitY and yitS, whose products respectively display significant homology with L-gulono-gamma-lactone oxidase of rat and dihydrofolate reductase of Staphylococcus aureus. The data also indicated that nprB mapped to a different position than previously published.
 
Transmission electron micrographs of the ancestor and two evolved lines, L16 and L18. (a, b) Ancestor. (c) Evolved line L16. (d) Evolved line L18. Bars, 1 µm.  
For legend see facing page.  
Clustering based on catabolic profiles of the 18 evolved lines, their two proximate ancestors that differ in antibiotic resistance markers (ANC-Str R and ANC-Nal R ), and the original Ralstonia strain TFD41 (ANC). The tree was constructed by hierarchical cluster analysis based on the method of Ward (1963), adjusted for covariance, using SYSTAT v. 7.01. See Table 1 for strain identification. ANC indicates the original Ralstonia strain TFD41, while ANC-Str R and ANC-Nal R denote antibiotic resistance marker variants thereof.
Cluster analysis of FAME data from the ancestral strains and ten evolved lineages. L1, L5, L6, L8, L9, L11 and L12 evolved in liquid medium, whereas L16, L17 and L18 evolved on agar. Each strain was tested in triplicate (denoted a, b and c). ANC indicates the original Ralstonia strain TFD41, while ANC-Str R and ANC-Nal R denote antibiotic resistance marker variants thereof.
Evolutionary pathways open to even relatively simple organisms, such as bacteria, may lead to complex and unpredictable phenotypic changes, both adaptive and non-adaptive. The evolutionary pathways taken by 18 populations of Ralstonia strain TFD41 while they evolved in defined environments for 1000 generations were examined. Twelve populations evolved in liquid media, while six others evolved on agar surfaces. Phenotypic analyses of these derived populations identified some changes that were consistent across all populations and others that differed among them. The evolved populations all exhibited morphological changes in their cell envelopes, including reductions of the capsule in each population and reduced prostheca-like surface structures in most populations. Mean cell length increased in most populations (in one case by more than fourfold), although a few populations evolved shorter cells. Carbon utilization profiles were variable among the evolved populations, but two distinct patterns were correlated with genetic markers introduced at the outset of the experiment. Fatty acid methyl ester composition was less variable across populations, but distinct patterns were correlated with the two physical environments. All 18 populations evolved greatly increased sensitivity to bile salts, and all but one had increased adhesion to sand; both patterns consistent with changes in the outer envelope. This phenotypic diversity contrasts with the fairly uniform increases in competitive fitness observed in all populations. This diversity may represent a set of equally probable adaptive solutions to the selective environment; it may also arise from the chance fixation of non-adaptive mutations that hitchhiked with a more limited set of beneficial mutations.
 
Top-cited authors
Ada Viterbo
Enrique Monte
  • Universidad de Salamanca
Ilan Chet
  • Hebrew University of Jerusalem
Rosa Hermosa
  • Universidad de Salamanca
Janet K Jansson
  • Pacific Northwest National Laboratory